Get the app
← back to catalog · registered 2026-08-22 13:56

josephmayo/Mellum2-12B-A2.5B-Thinking-Abliterated-GGUF

josephmayo 12B GGUF MoE second-order 131K ctx
Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model. 

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/josephmayo%2FMellum2-12B-A2.5B-Thinking-Abliterated-GGUF"
Response includes
  • classification m8
  • files 5
  • hub_downloads_all_time 776
  • author_summary 4 models
  • readme_text full
Try next
GET/models?author=josephmayo
All 4 models by josephmayo
 GET/methods/m8
What "m8" means · taxonomy
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M8
Primary method

Repackaging (quantization)

Applied on top of direct removal inherited from the base model.
Confidence
MEDIUM
Inherited from base model
M1 Direct removal →
Why this label 3 signals
Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • 'abliterated' in name/tags
  • is_gguf=1
  • assume M1 (base ablation) + M8 (GGUF quant) - default when producer unknown
How we label → About M8 →
Refusal direction extraction

No specific extraction method could be identified for this model. The producer either did not document it or used a proprietary pipeline.

What is a refusal direction? →
Downloads · lifetime
776
662 last 30d - active
Likes
1
Model age
4mo ago
created 2026-06-02
Downloads over time
Now1.3K→from396↑238%
3497101.1K1.4K396 on Jun 101.3K on Oct 11JunJulAugSepOct
Jun 10 → Oct 11 · 57 snapshots · spans 123 days

Genealogy 0 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

josephmayo/Mellum2-12B-A2.5B-Thinking-Abliterated↓ 31 this lineage
josephmayo/Mellum2-12B-A2.5B-Thinking-Abliterated-GGUF↓ 662

Variants by this author 2 formats · 693 downloads combined

The same weights this author released in different packaging. Pick the format that matches your runtime.

Metadata

Quantizations
Q4_K Q6_K Q8_0
Tags
gguf abliterated mellum2 moe text-generation base_model:josephmayo/Mellum2-12B-A2.5B-Thinking-Abliterated base_model:quantized:josephmayo/Mellum2-12B-A2.5B-Thinking-Abliterated endpoints_compatible region:us

Related

Other by josephmayo
Fara-7B-Abliterated-v2-GGUF 156 Mellum2-12B-A2.5B-Thinking-Abliterated 31 Fara-7B-Abliterated-v2 22
Same base model
mradermacher/Mellum2-12B-A2.5B-Thinking-Abliterated-i1-GGUF 995 mradermacher/Mellum2-12B-A2.5B-Thinking-Abliterated-GGUF 440
Total size
29.8 GB
Files
5
Quantizations
4
Registered
2026-08-22 13:56
Last updated on HF
2026-06-02 23:08
Source huggingface.co/josephmayo/Mellum2-12B-A2.5B-Thinking-Abliterated-GGUF ↗ Base model josephmayo/Mellum2-12B-A2.5B-Thinking-Abliterated →

Files by quantization

Q8_0 1 file 12.0 GB
Mellum2-12B-A2.5B-Thinking-Abliterated-Q8_0.gguf 12.0 GB 40a96b45 download
Q6_K 1 file 10.1 GB
Mellum2-12B-A2.5B-Thinking-Abliterated-Q6_K.gguf 10.1 GB c64452cf download
Q4_K 1 file 7.67 GB
Mellum2-12B-A2.5B-Thinking-Abliterated-Q4_K_L.gguf 7.67 GB 5519d500 download
Auxiliary files 2 files 2.52 KB
.gitattributes 1.73 KB 0c99dc80 download
README.md 804 B ae2dc510 download

README current version from Hugging Face

base_model: josephmayo/Mellum2-12B-A2.5B-Thinking-Abliterated
library_name: gguf
pipeline_tag: text-generation
tags:

  • gguf
  • abliterated
  • mellum2
  • moe

Mellum2-12B-A2.5B-Thinking-Abliterated - GGUF

GGUF quantizations of josephmayo/Mellum2-12B-A2.5B-Thinking-Abliterated,
an abliterated build of JetBrains Mellum2-12B-A2.5B-Thinking (64-expert MoE, top-8 active).

Produced with llama.cpp (commit 63e66fdd23eda3a2659a7af9ff6ef15d71efbff1), static quants (no imatrix).

Quant Notes Size (GB)
Q8_0 standard 12.93
Q6_K standard 10.88
Q4_K_L Q4_K_M body + Q8_0 embed/output (bartowski-style) 8.24

Disclaimer

Research artifact only. Not for harmful use.

README history 1 version

The author's README evolved over time. Click a version to see its content at that point.

  1. 2026-06-02Upload README.md with huggingface_hub199b6f1804 B
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Get the app Browse the catalog
Open in Abliteration